Marine coastal habitats in the Mediterranean Sea are severely biologically invaded ecosystems (Boudouresque and Verlaque, 2002). This is largely due to human-mediated introduction of non-native species through various vectors such as ballast waters of ships, regional aquaculture activities, and by natural dispersal via the Suez Canal (Lessepsian invasion) (FD, 1978; Grosholz, 2002; Rilov and Crooks, 2009; Katsanevakis et al., 2014). The total number of non-indigenous seaweed species (NISS) reported for the whole Mediterranean Sea is somewhat variable, which is not surprising given the complex cryptic nature of some of the species involved. The date of introduction and total numbers are subject to conjecture; nonetheless, agreed estimates may vary from as high as 148 species (as calculated for the year 2022 using the suggested 2–3 new immigrants per year by Zenetos et al. (2012), down to 118 species (Verlaque et al., 2015). The arrival of NISS is more intense in the eastern Mediterranean Sea since there is a consensus that the major vector of the introduction of macroalgae is of Lessepsian origin (Boudouresque and Verlaque, 2005; Zenetos, 2010; Zenetos et al., 2012; Verlaque et al., 2015; Romero et al., 2016).

In the last decades, the Israeli Mediterranean coast has become the host of several NISS, primarily in subtidal areas, and witnessed intense onshore drifts (Israel et al., 2019). Examples are the green alga Codium parvulum (Bory ex Audouin) P.C.Silva (Israel et al., 2010), the red alga Galaxaura rugosa (J.Ellis & Solander) J.V.Lamouroux (Hoffman et al., 2008), the brown algae Stypopodium schimperi (Kützing) Verlaque & Boudouresque (Verlaque and Boudouresque, 1991; Einav and Israel, 2009), Dictyota sp. (unpublished observations) and Lobophora lessepsiana C.W.Vieira (Vieira et al., 2019), and many others which are unaccounted for. In this context, the red alga Asparagopsis taxiformis (Delile) Trevisan, allegedly introduced into the Mediterranean Sea in 1831 (Verlaque et al., 2015) was hardly noticeable in the Israeli shores until about a decade ago. A. taxiformis is highly invasive and has therefore been underpinned as one of the ‘worst invasive alien species threatening marine biodiversity in Europe’ (EEA, 2007), and further listed within the 100 ‘worst invasive seaweeds in the Mediterranean Sea’ (Streftaris and Zenetos, 2006)

A. taxiformis is regarded as a cryptic complex with high diversity and cosmopolitan distribution from warm-temperate to tropical marine environments (Harvey, 1849; Price et al., 1986; Boni and Hawkes, 1987; Silva et al., 1996; Zanolla et al., 2022). The heteromorphic life cycle of A. taxiformis includes the erect gametophyte and a filamentous tetrasporophyte (Mairh, 1977; Guiry and Dawes, 1992). The gametophytic stage is highly branched, colored pink to red reaching up to 40 cm tall and commonly occurs on rocky substrates, or as epiphytes (Rojas et al., 1982; Boni and Hawkes, 1987). The tetrasporophytic stage, mistakenly referred in the past as a different species (Falkenbergia hillebrandii (Bornet) Felkenberg) (Chihara, 1961), consists of microscopic three-cell row filaments arranged in a pompon morphology up to 2 cm in diameter. Tertrasporophytes can be found free-floating or attached to other algae and are capable of dispersion by flotation. The high vegetative reproduction potential of the tetrasphorophyte stage serves as a prolific propagator for expanding population and invasion to new habitats (Chihara, 1961; Mairh, 1977; Zanolla et al., 2022).

To date, worldwide, six mitochondrial lineages have been genetically distinguished in the A. taxiformis populations (Andreakis et al., 2007; Sherwood, 2008; Dijoux et al., 2014; Andreakis et al., 2016; Zanolla et al., 2022). The cox2-3 spacer is considered the ideal marker for this linage separation differing in 6-23 bp out of the 338 bp of cox2-3 spacer marker (Dijoux et al., 2014; Andreakis et al., 2016; Kurihara et al., 2016). Based on this molecular tool, lineage 1 (L1) was described for the Pacific region, whereas a L2 additionally described for the Indo-Pacific (Andreakis et al., 2007; Dijoux et al., 2014), the Mediterranean Sea (Andreakis et al., 2007), and North Atlantic regions (Dijoux et al., 2014). L3 is known for the Mediterranean Sea (Andreakis et al., 2007) and also described for the western Atlantic (Andreakis et al., 2007), the Canary Islands and South Africa (Bolton et al., 2011). L4 can be found in Hawaii and the western Indo-Pacific (Sherwood, 2008) while L5 is found in Western Australia and the southern Indo-Pacific Ocean (Dijoux et al., 2014; Andreakis et al., 2016). Recently, L6, considered endemic to Australia was described by Andreakis et al. (2016). As mentioned before, A. taxiformis was first documented in the Mediterranean Sea before the opening of the Suez Canal, and its current distribution is likely the result of several introduction events including possible Lessepsian migration (Andreakis et al., 2004; Ní Chualáin et al., 2004; Andreakis et al., 2007; Dijoux et al., 2014). To date, only two cryptic lineages, L2 and L3, have been described for the Mediterranean Sea (Zanolla et al., 2022). Within the Israeli Mediterranean Sea (IMS), A. taxiformis has become abundant at 0-15 m deep, generally attached to hard bottoms. The presence of A. taxiformis along the Israeli coasts has been documented only twice (Lipkin, 1962; Einav and Israel, 2008), and no ecological or molecular biodiversity were reported for the local populations in the past. Field collections of specimens preserved at the Seaweed Herbarium of the Israel Oceanographic & Limnological Research, Ltd. (IOLR) (www.seaweedherbarium.com) confirm the expansion of this species during the last decade.

Global human activities are responsible for the spreading of marine organisms and the changes imposed on marine ecosystems on a worldwide scale. To detect and follow those changes, surveys on a local scale are necessary, having significant implications both for marine conservation and coastal ecosystem management (Bickford et al., 2007; Andreakis and Schaffelke, 2012). In this study we investigate for the first time the A. taxiformis populations of the IMS, integrating a DNA barcoding approach with morphological tools. In order to gain a better understanding of the distribution and origin of the IMS A. taxiformis, three genetic markers, nuclear LSU, mitochondrial cox2-3 spacer and plastid rbcL were used.

Overall, 30 specimens of Asparagopsis taxiformis, 28 gametophytes and 2 tetrasphorophyte were collected in five sites along the northern Israeli coastline, from July 2020 until December 2021. Sampling efforts in southern shores, which are largely composed of sandy bottoms, indicated the absence of this species in soft bottoms. Morphological features of mature samples collected from the dense Rosh Hanikra site populations are shown in Table 2. IMS samples showed cystocarps (female reproductive structures) and spermatangia (male reproductive structures) in the majority of the samples examined (Figures 2, 3). Specifically, out of the 28 collected gametophytes, 20 were found to be monoecious, and no reproductive organs were observed for the rest.

A total of 27 sequences were successfully generated for the cox2-3 spacer marker, 29 for the rbcL, and 23 for the LSU (Table S1). Phylogenetic trees constructed for the three different genes (Figure 4) revealed that all samples examined clustered within the lineage 4 (L4). All three markers, cox2-3 spacer, LSU and rbcL, show clustering with the IMS samples together with all L4 lineages samples deposited from GenBank. One exception is the LSU marker of the L4 sample (GenBank KJ772105) from the Egyptian Red Sea which clustered alone and may be a result of a single nucleotide polymorphism which is unique from all of the other samples and most likely may represent a sequencing error. Previous sequencing of the cox2-3 spacer from distant geographic sites such as Hawaii and Panama differ by 0.98% (3 out of 306 bp) from the IMS samples. Closer sequencing can be seen from L4 samples from French Polynesia and Papua New Guinea which have a discrepancy of 0.65% (2 out of 306 bp) from the IMS samples (Andreakis et al., 2007; Sherwood, 2008; Bolton et al., 2011; Dijoux et al., 2014; Andreakis et al., 2016; Kurihara et al., 2016). Samples from South Africa, Europa Island, and Sri Lanka diverged from the IMS samples in only one nucleotide (0.32%). The closest sequencing to the IMS samples is A. taxiformis L4 sampled from the Egyptian Red Sea with 100% similarity (Figure 5). In comparison to previously sequenced Mediterranean samples, the IMS cox2-3 marker differs by 2.3% and 3.9% from L3 and L2 Mediterranean populations, respectively (Andreakis et al., 2004; Andreakis et al., 2007; Dijoux et al., 2014) (Figure 5).

The seaweed Asparagopsis taxiformis is an iconic example of macroalgal invasion from a global perspective, which has invaded almost all oceans and seas around the world (Zanolla et al., 2022). Until now, two cryptic lineages have been described for the Mediterranean Sea, L2 and L3 (Zanolla et al., 2022). In this study, we present solid evidence of an additional cryptic lineage (so-called L4) of A. taxiformis first reported for the Mediterranean Sea. We regard this lineage as relatively new when compared to the introduction of L2 and L3 genetic lines that were reported in the past. The wide geographic distribution of A. taxiformis L4 seems to be the result of long-distance, human-mediated dispersal events (Dijoux et al., 2014). The true native origin of L4 is unknown, however, the pathways of introduction into the Mediterranean Sea can be predicted. A soundly invasion route includes two distinct geographic origins, Sri Lanka or South Africa, both within the Indian Ocean. A. taxiformis populations may have established in the eastern Indian Ocean and reached the Red Sea, to eventually penetrate into the eastern Mediterranean Sea via Suez Canal. This possible route is supported by the similarity of the cox2-3 marker from the IMS sequences and between South Africa, Sri Lanka, and the Red Sea specimens (Figure 6). Whether A. taxiformis is a native species or has been introduced into the Mediterranean Sea is still unclear (Boudouresque and Verlaque, 2002; Zanolla et al., 2022). However, based on previous data collected on common Lessepsian migration route (Nunes et al., 2014) taken together with the proximity of the IMS to the Suez Canal, as well as the molecular data presented in this study, we hypothesize that A. taxiformis L4 is of a Lessepsian origin. From our observations and viewing the lack of ecological and molecular studies in the past, we suggest that A. taxiformis L4 introduction into the eastern Mediterranean Sea may have occurred only a few decades ago.

Morphological characters of A. taxiformis gametophyte have been used to distinguish between cryptic lineages (Zanolla et al., 2014). Furthermore, gametophytes from the same lineage may show different characters when comparing populations from different geographic regions. For instance, thallus height, basal diameter, and length of the branchlets were all greater in Mediterranean Sea invasive L2 as compared to Hawaiian L2 populations (Zanolla et al., 2014). The trend of increasing thallus morpho-parameters in the Mediterranean Sea populations compared to populations from the same lineage, but from distant geographic sites, was also seen in L4. Specifically, morphological traits of L4 from Rosh Hanikra site, namely thallus height, main axis basal diameter, and height where branching begins were higher compared to L4 from Hawaii. There is still uncertainty as to whether this tendency is due to the different biotic conditions in the Mediterranean Sea or whether the increased size of some specimens gives them ecological advantages, thus increasing their invasive rates and extent. Recently observed drifts strongly indicate that A. taxiformis is widespread in shallow habitats between Rosh Hanikra and Mikhmoret sites. In these sites, A. taxiformis is frequently attached to exposed rocky substrates both vertically and horizontally, usually between 0-5 m deep. Nonetheless, A. taxiformis is also quite abundant at depths of 20–30 m and significant drifts of fertile gametophytes are carried out to the shore following stormy sea days. We have no quantitative seasonal data for the above observations, yet our frequent field trips indicate that A. taxiformis exists all year round with productivity declining during wintertime. Growth capacity together with physiological and reproductive traits related to environmental factors for the IMS L4 A. taxiforims are still unknown. In a previous study which did not include genetic characterization, some populations of A. taxiformis were described as dioecious (Barone et al., 2013). As for the L4 lineage from Hawaii, Zanolla et al. did not report the presence of any reproductive structures (Zanolla et al., 2014). In this study, nearly every specimen showed reproductive structures carrying both cystocarps and spermatangia strongly indicating that IMS L4 is monoecious (Figure 3).

Detached gametophytes as seen in large amounts in the IMS could promote further dispersal of the L4 population in two possible ways. First, by vegetative reproduction via fragmentation of the thalli as suggested by Mairh (1977) and second, although the released gametes and tetraspores are not thought to travel very far (Santelices, 1990), detached gametophytes that drift with the currents can carry and release reproductive structures for long distances. The presence of the gametophyte in a site does not necessarily confirm the presence of the tetrasporophytes stage, and vice versa (Bolton et al., 2011; Orlando-Bonaca et al., 2017). The observation of both reproductive stages in the same site indicates an active reproductive population. Furthermore, the occurrence of tetrasporophytes, which are known as a good dispersal unit, makes this population a potential donor source for future invasive populations

All A. taxiformis specimens analyzed from the Mediterranean Sea so far were found to be either L2 or L3 (Zanolla et al., 2022). Studies involving the population dynamics of the Mediterranean L2 suggest the successful expansion of this lineage can be explained by the temperate and tropical preferences of this lineage (Zanolla et al., 2018). Furthermore, rising sea temperatures in some regions due to climate change will also increase the opportunity for this lineage to further spread in those areas (Zanolla et al., 2018). Zanolla et al. (2018) also highlight the co-existence of L2 and A. armata in some areas, suggesting as the former existence of A. armata promotes the L2 expansion in those sites. This conjecture relies on the assumption that communities under stress from an invasive species are more susceptible to additional invaders (Simberloff and Von Holle, 1999). Considering that A. taxiformis is already well established in almost all of the Mediterranean Sea, the ability of L4 the flourish in tropical-temperate regions, and the suggested Lessepsian migration route, there is a significant likelihood that L4 populations will spread to other Mediterranean basins. Moreover, considering the classic Lessepsian migration route of NISS we hypothesize Tsiamis and Panayotidis (2007) first description of A. taxiformis in Greece (which not include molecular work), might have corresponded to the L4 described in this study. The increasing abundance of A. taxiformis L4 in the eastern Mediterranean Sea underlines the need for a comprehensive monitoring strategy for its distribution and lineage description. Lastly, analyses of macroalgae herbaria samples may be crucial to assess their impact on local biodiversity and track their possible route over time.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

All authors design the research and conduct the field collection. ON did the molecular work and data analysis. All authors provided critical feedback and helped shape the research, analysis, and manuscript. All authors read and approved the final manuscript.

This project was funded by the Ph.D. excellence scholarshipof the University of Haifa given to ON, and by Ministry ofHealth, Israel grant no. 3-16052 given to AI.

We are thankful to Guy Paz for preparing the artwork. In addition, we want to thank Dr. Maya Lalzar from the University of Haifa Bioinformatics Service Unit for her assistance and Dr. Dikla Aharonovich for her help with electron microscopy.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.